Online negotiation and mediation: Is there room for arguement support tools? Some re... Page 1 of 10 17th BILETA Annual Conference April 5th - 6th, 2002. Free University, Amsterdam. Online negotiation and mediation: Is there room for argument support tools? Some reflections on argumentation and eadr Arno R. Lodder (Computer/Law Institute, Vrije Universiteit Amsterdam) Abstract Alternative Dispute Resolution (negotiation, mediation, arbitration) is an alternative to traditional dispute resolution, in particular litigation. In our present Information Society the shift to the online environment has also taken place for ADR. This new field is usually called ODR: Online Dispute Resolution (Katsh & Rifkin 2001). Parties in an `alternative' dispute (e.g. ODR) do not primarily try to fight each other as opposing parties, but rather aim at finding a solution for their problem cooperatively. This consensual characteristic makes that at first sight argument support models do not seem to be suited for eadr, because these models are adversarial based. Lodder & Huygen (2001) describe an argument support tool that was primarily meant to support the parties in online domain name disputes. They claimed the tool could also be used in online negotiation and mediation. This paper provides a preliminary analysis on whether argument support tools can be used in online mediation and negotiation. 1. Introduction Nowadays it is rather easy to order products from anywhere around the world. If the ordered product is delivered in time and undamaged, and the buyer is satisfied in all other respects, it makes no difference in whatever country the seller is established. But, what if something goes wrong, as in the following example? The Dutch student John Patrick ordered a CD-ROM player from an Italian company, via their site www.cdrom4sale.com. When the CD-ROM player is delivered, John finds out that different from what he expected he has to pay EURO 200. Beside EURO 180 for the CD-ROM player, he also has to pay EURO 20 for delivery costs. John is surprised, because when he ordered the CD-ROM player it was not indicated that he had to pay delivery costs. The Italian company argues that it speaks for itself that the deliverance of a CD-ROM player costs money, reason why they do not indicate on their website these costs. Those who are familiar with e-commerce regulations, will recognize that the Italian company has
Online negotiation and mediation: Is there room for arguement support tools? Some re... Page 2 of 10 violated both Article 4(1)(d) of Directive 97/7/EC on Distance Selling and Article 5(2) of Directive 2000/31/EC on Electronic Commerce. If conflicts arise, the fact that the seller is established in a foreign country can become a problem. You cannot visit the seller and talk things out, and making a long distance phone calls is quite expensive. Going to court is not an option either, since the costs involved in a court procedure would be much higher than the damage suffered. Online dispute resolution (ODR), in particular if there are no costs involved, could be a way out here (Bordone 1998). This paper is about ODR, and explores in what way argument support tools might be used in online negotiation and mediation. The paper is structured as follows. Section 2 briefly introduces the notion eadr. Subsequently in section 3 an argument support tool is described, and in section 4 a sample Smartsetlle negotiation case is discussed. Section 5 discusses requirements for argument support tools in online negotiation, and in section 6 concluding remarks are made. 2. eadr or ODR Although still in its infancy, ODR could become an important way to solve conflicts online. ODR is the online version of Alternative Dispute Resolution, basically divided into negotiation, mediation, and arbitration. In case of negotiation parties negotiate in order to resolve a conflict or to work out a compromise, without a third party being involved. In case of mediation an independent third party, the mediator, moderates the discussion between the parties. However, the mediator does not have the power to impose a solution upon the parties. The influence of the third person is most prominent in case of arbitration. Like in litigation, the arbiter decides the case. As an explanation for the fact that `alternative' is omitted from the term ODR, Colin Rule has claimed that: "For the first time, dispute resolution isn't really an alternative--the courts don't work online, so dispute Resolution is often the default." Although right at that moment (Fall 2000), currently some courts are actually going online (e.g. in the US state Michigan). There is, however, another reason why ADR is called alternative, namely, the approach to dispute resolution is different. Parties in an `alternative' dispute do not primarily try to fight each other as opposing parties, but rather aim at finding a solution for their problem co-operatively. This is a reason why some people prefer the term eadr, where the `e' refers to electronic like in ecommerce, eeurope and edirectives (Lodder & Kaspersen 2002). 3. The eadr tool 3.1 Theoretical results and practical applications In (Lodder 1998) the following observation was made: "In the AI & Law field many dialogical models have been developed. Since legal practice is governed by procedures these AI & Law models are suited to adapt in order to be used by legal professionals and/or paralegals. The prospects are good, regarding automation in the legal field. A necessary condition for dialogical models to be used in practice is that the resulting dialogs are natural. (...) Future research will also concern how exactly the existing models should be adapted to serve a useful role in practice. This research is both interesting and promising. We should try to achieve that lawyers are using applications in which results from AI & Law research are integrated." It was not until last year that I developed a model that was meant to met the above requirements (Lodder & Huygen 2001). 3.2 Introductory observations
Online negotiation and mediation: Is there room for arguement support tools? Some re... Page 3 of 10 In the last ten years quite some AI & Law research was dedicated to the development of dialogical models, also called dialog games (for further reading: Bench-Capon 2000, Gordon 1995, Hage 2000, Leenes 1999, Lodder 1999). One of the purposes of these models is to structure the argumentation of the users of the model. The purpose of the present model is to structure the information of the parties in online ADR. Our model is not meant to replace existing ADR procedures, but to be incorporated into them. A dialog game can be used by a single player, to test the acceptability of his statements critically, and to adduce as many supporting statements as he believes necessary to convince an imaginary opponent. However, the idea underlying the present model is, that a party hands over the result of a single player game to his opponent. In case of online arbitration, the arbiter can use the model to gather the information necessary for deciding the case. 3.3 Defining the model The central elements are the parties, the statements and the games' board. They are defined as follows. THE PARTIES There are two parties, the Complainant and the Respondent. STATEMENTS A statement is an expression in natural language. Each statement deals with only one topic. An issue is a statement. Supporting statements support an issue. Attacking statements attack a statement. THE GAMES' BOARD The games' board is empty at the first turn. The games' board consists of all statements added by the parties according to the rules. The parties move alternately, exchanging after each turn the games' board. The Complainant and Respondent can introduce statements according to the rules that are discussed below. Statements are not required to be cast in a prescribed format, but they can be put forward in natural language. The only requirement is that a statement addresses a single topic. There are three types of statements. Besides statements that formulate issues, there are statements that support issues and statements that attack other statements. The games' board is a metaphor, used to describe the collection of statements added by the parties. In most existing dialog models, parties can perform four types of moves: claim, accept, question and withdraw. However, by only allowing parties to claim statements, the information of the parties in online arbitration can be structured sufficiently with two simple rules. RULE 1 - ISSUES AND SUPPORTING STATEMENTS First step: Introduce an issue Second step: Adduce one or more statements supporting the issue (supporting statements). RULE 2 - RESPONDING STATEMENTS In reaction to any statement of the opponent, one or more statements can be put forward (responding statements). The first rule is about the introduction of issues and supporting statements, the second rule about the reaction to statements. These rules have been used to define three models: the basic model, the 2-
Online negotiation and mediation: Is there room for arguement support tools? Some re... Page 4 of 10 turn-issue model, and the n-turn model. Only the general n-turn model is discussed here. THE N-TURN MODEL First turn: Complainant executes rule 1 as often as necessary. Consecutive turns: The parties execute rule 2, and execute rule 1 as often as necessary. In the first turn, the Complainant executes rule 1. He starts with identifying an issue (first step). Subsequently he adduces statements supporting the issue (second step). Note that a statement deals with only one topic, reason why he can enter one or more supporting statements. Finally, the Complainant applies the two steps of rule 1 again if there are more issues to be identified. If the Complainant is finished, he hands over the games' board. In consecutive turns the parties can react to all statements they do not agree with. Note that an issue is a statement, so the parties can also react to issues. At each turn the parties can introduce new issues. It might be that providers of ODR want to restrict the introduction of issues. Therefore, and in line with civil proceedings, the not discussed 2-turn issue model restricts the introduction of issues to the first two turns. 3.4 The implementation of the model Below is an example of the implemented model. The above figure is a screen-shot of the odr applet, running in a Netscape browser. It represents a domain name case (Davidgilmour.com). A full analysis of the case can be found in (Lodder & Huygen 2001). Each statement occupies a line and the indentation indicates the structure of the statements in the argument. The color indicates whether a statement is an issue, a supporting or an attacking statement. When the user clicks on a statement with the mouse, she can read the full text of the statement, and she can attach another statement to it, if she is entitled to do so. 3.5 Evaluation The above model could prove helpful when used in an online domain name dispute. First, the parties do not have to use any formal language, but in stead the tool structures the natural language input, and helps the parties to focus on one topic at the time. As a result, the parties can communicate all relevant information in a format that facilitates to quickly understand what exactly the dispute is
Online negotiation and mediation: Is there room for arguement support tools? Some re... Page 5 of 10 about. So, the arbiter or the opponent does not receive a rather unstructured, running text, but in stead structured information that shows all issues and statements supporting the issues, as well as any responses to these statements. Second, if the case is published on the website of the arbiter in a structured format, searching prior verdicts would become easier than it is nowadays. For example, by using the structure of the documents, one could easily retrieve all verdicts in which the respondent did react, and, if so, to what particular issues he did react. In domain name cases often the respondent does not react at all. In the case presented above, the respondent did react albeit not to all statements. It is not part of the model yet, but as someone has suggested to me it would be a good idea if the arbiter would in turn use the tool to enter comments and, also, his decision. In domain name cases there are three conjunctive issues. So, in order to rule in favor of the complainant (which mostly is the case in these domain name, anti-domain name grabber disputes) the arbiter must indicate his approval for each of the issues. Whereas the online domain name arbitration procedure is really suited for the discussed tool, Tom Gordon suggested that the tool might turn out to be not that helpful in negotiation and mediation. The reason for his suggestion was that the philosophy of ADR is based on co-operation, and argument support tools are adversarial based. The following quote by Thiessen & Macmahon (2000) emphasizes that adversarial tactics are usually counter-productive: "competition in negotiation often results in harmful adversarial tactics, and inefficient results." So, at first sight argument models do not seem to be suited for negotiation and mediation. Lodder & Huygen (2001) claimed that the above tool could also be used in online mediation and negotiation. It may seem questionable whether this claim can be substantiated. However, I do believe that the argument tool could be used in negotiation and mediation. In particular the tool could be used in combination with existing negotiation and mediation tools. To illustrate this, I will first discuss a sample negotiation case in which negotiation software was used, and indicate in what way the argument tool could prove useful during these negotiations. Next I will discuss some further requirements argument support tools should meet when used in negotiation or mediation. In the remainder of this paper I will primarily use negotiation/facilitation as an example. The special role of the mediator is not taken into account, but most remarks regarding using an argument support tool in negotiation are equally applicable to mediation. In the end mediation is a type of negotiation, where a mediator actively steers the negotiations both procedurally and regarding the content. 4. A sample negotiation case Two parties are involved in the sample case I will discuss: DEC and Riverside. The case was presented in Thiessen & Macmahon (2000), and is an adaptation of a "DEC v. Riverside Lumber" as developed by Lax, Sebenius, Susskind & Weeks for the Harvard Law School Program on Negotiation. The software used by the parties in the dispute is SmartSettle (for a comparable AI & Law tool, see Belucci & Zeleznikow 1999), a sophisticated tool that can be used in multi-issue, n- party negotiations. I will not repeat the complete negotiations between DEC and Riverside, but emphasize the main points. The background of the conflict is as follows. Riverside is located beside the Deep River, which happens to be a convenient place to discharge the effluent from their operation. The Deep River water quality has deteriorated over the years and an administrative order has been issued by DEC requiring Riverside to improve the situation. Because a demonstration project would be of great value, DEC is quite keen on having Riverside install a new technology called the Technoclean Scrubber. DEC is so motivated toward the Technoclean Scrubber that it seems willing to include a guarantee, subsidy, insurance, and other incentives. However, DEC also claims the authority to require Riverside to shut down for as long as it takes to properly comply
Online negotiation and mediation: Is there room for arguement support tools? Some re... Page 6 of 10 and go through a testing period. Not surprisingly, Riverside is unwilling to cooperate with DEC's ideas for resolution of the conflict. (Thiessen & Macmahon 2000) The issues in the dispute are: The scrubber type Guarantee for the scrubber Subsidy Insurance Shut down PR Budget Other incentives After identifying the issues, parties exchange proposals in which each issue is given a value. The negotiations end if both parties accept the same proposal. The software also makes proposals on the basis of the input (values, bargaining ranges) of the parties. In the current dispute the scrubber type is a bivalent issue: Riverside prefers the proven technology of Rotoblue, while DEC wants the new Technoclean. The guarantee issue is bivalent too (yes or no). Subsidy, insurance, PR budget and other incentives are financial issues. The value of the shut down period is also a numerical issue, viz. the number of months. Surprisingly, contrary to what the parties actually want, in the initial proposal DEC asked for no shut down of Riverside, while Riverside proposed a shut down period for 2 months (see the table below). This tactic is probably used by the parties to please their opponent. For Riverside there appeared to be an additional reason, namely that they could well use a shot down period for internal reorganization. Obviously, this information is not shared with DEC. It provides, however, support for the position of Riverside. Since SmartSettle has private sections in their software, Riverside could use the argument support tool to express the support for this position. In particular since the position is contrary to what one would expect, the storage of this information is useful both for the current negotiations and for any future use of the results of the negotiations. In case a facilitator (like SmartSettle does) or mediator is involved this information could be shared in the caucus. An example of support information that could be shared with DEC is regarding the guarantee. DEC is offering to include a guarantee for the scrubber, the reason Riverside does not need a guarantee is that the Rotoblue scrubber is based on proven technology. A really advanced application of the argument support tool would be if explanations for proposals made by the SmartSettle software would be automatically generated. These explanations might also contain secret information, namely information that explains the proposal but is based on information only known to a single party. This could mean that a provided explanation is insufficient, namely if information that is necessary for an explanation is only known by the other party. The following table show some rounds of this negotiation. R stands for Riverside, D for DEC. The numbers stand for thousand dollars, unless indicated otherwise. ISSUE BATNA of R Initial proposal Second proposal SmartSettle: Equivalent R D R D SmartSettle: Improve Scrubber Tech. Roto. Tech. Tech. Tech. Tech. Roto
Online negotiation and mediation: Is there room for arguement support tools? Some re... Page 7 of 10 Guarantee No No Yes Yes No Yes No Subsidy 100 0 10 100 100 89.7 24 Insurance 100 200 100 150 150 150 90 Shut down 6 months 2 months No No No No 2 months PR budget 50 100 50 30 30 49.5 130 Incentive 100 150 100 150 150 130.5 80 Total -588-362 -655-545 -592-545 -455 Table - The Riverside v. DEC dispute The total at the bottom of the table represents the number the SmartSettle software uses to calculate the value of the different proposals. I am not sure what value the non-financial issues represent, but these are probably "translated" into financial issues. The equivalent proposal differs on the various issues from the second proposal by Riverside, but the total amount is the same (equivalent). This equivalent proposal would represent from the perspective of DEC 592 points. So, the same proposal is calculated differently for each party. For example, the scrubber Technoclean is of more importance to DEC than it is to Riverside. The automated generations of explanations might help here in particular to explain to both parties why the last proposal made by SmartSettle is an improvement. As already indicated above, the explanation will not necessarily be the same for both parties. It is interesting to note that the money involved in the second proposal is the same for both parties: Riverside would get $ 430. In the equivalent proposal the total amount money has dropped to $ 419,7, and in the improved proposal the total has even further dropped: $ 324. In the last proposal the PR budget has been raised to an enormous amount, even far beyond any proposal by the parties. Again, in particular regarding the PR budget some explanation would be helpful: automatically generated explanations presented in the argument support tool. 5. Argument support tools in online negotiation: requirements A fundamental difference between negotiation and models of argumentation is that the latter are primarily focused on obtaining your position, while the aim of the former is "not obtaining your position but satisfying your interests." (Ury 1991, p. 18) A position is related to an issue in a dispute. If parties share a position, there is agreement on the issue. Positions might change during negotiations (e.g. due to bargaining), interests do not. The following family law example illustrates the difference between positions and interests. Assume that Mary and Barry have been married for 10 years and decide to get a divorce. They own a $ 300.000 house, and a car worth $ 50.000. If Mary wants the car and Barry wants the house, these desires represent positions. The interest of each of the parties is less concrete, and in case of a divorce probably will be to get an even share (or, sadly, might be to annoy and frustrate the other as much as possible). At first sight it seems that an interest is not worth arguing about, and a position is. An interest seems to be an argument in itself, interests seem self-evident. However, a single party can have different, sometimes even contradicting interests. In that case support for these interests might help in getting the dispute solved. For example, an employee may want to earn as much money as possible on the one hand, and work only as little hours a week as possible on the other hand. Undeniably, in case of such contradicting interests satisfaction of one interest will lead to dissatisfaction of the other. Normally, justifying interests will not help the negotiations between for example this employee and his employer when discussing the salary and working hours. Adducing support for a position may not help either. For example, in what way could support for the position of the employee that he wants to work 3 days a week and get paid $ 5000 a month help to reach agreement? Nonetheless, the
Online negotiation and mediation: Is there room for arguement support tools? Some re... Page 8 of 10 employee may provide an explanation, or the employer might ask for one. Providing a justification or giving an explanation, means supporting a position or an interest. The argument support tool can be used to register the support of a position or interest. In the current example the employer might have to spent time at home due to family circumstances and need the money for the same reason. This information supports both his interests and his positions. So, no matter whether during this negotiation parties exchanged positions or interests, the argument support tool could be used. As a consequence, the employer could be willing to meet the demands of the employee, for example during the period the family circumstances remain the same. The crux of the above example is to show that argument tools might be helpful in negotiations. In general, an explanation or justification can make the other party understand what point, and in particular why, the opponent is making. What is important is that the tool is used in spirit with the general philosophy of ADR: cooperatively. So, the tool must not be used to simply throw in ones positions provided with support, and, in case the other party reacts to the positions or the supporting statements, just adding more support and attacking the reaction of the opponent. It is important that parties try to listen to one another and try to understand each other. Using only an argument support tool may lead to two fighting parties, each just arguing for its own position. Although parties which proceed from the assumption that the negotiations will only be successful if they stand aside each other rather than in front of each other may be successful when only using the argument support tool, the use of the tool in combination with other ADR software should be preferred. The support tool could for instance be used in combination with the SmartSettle software. As noted at the beginning of this section, obtaining positions is not what negotiations should be about. It appeared, however, that arguments put forward to justify positions or interests might help to solve the dispute. Showing the reasons for your position not necessarily leads to adversarial discussions. Rather, providing explanation or justification can lead to understanding, so in stead satisfies the co-operative goal of ADR. One has to bear in mind though that if an argument support tool is used, it has to be used cautiously and not to just support ones position in order to obtain whatever is desired. A problem with using an argument support tool in negotiations is that not under all circumstances providing support is significant. Support can be trivial, e.g. if a party in a divorce case wants to have a painting made by his grandmother and both former partners know how important this painting is to him/her, providing support would not have any value. Also, support might reveal too much about your bargaining position. If in the SmartSettle case Riverside would support the statement about the 2 months closing, this would weaken their bargaining position. This and the other example illustrates that using the argument support tool must be optional, not obligatory. So in order to become a successful part of online negotiations, the argument support tool described should be adapted. The second step of rule should be optional. A purpose for which the argument support tool could be used is to identify at each moment during negotiations on what issues parties agree. In case of numerical issues disagreement on the amount not necessarily means disagreement about the issue itself. For instance, if a party is willing to compensate $ 100 and the other party wants $ 150 then there is consensus regarding the compensation, the dispute only concentrates on the amount. In the SmartSettle-example there is consensus on insurance, incentives and PR, but the proposal lacks consensus on subsidy. The argument support tool could be used to make this clear. In the proposal DEC wants to pay subsidy, but Riverside does not want to be subsidized. One cannot conclude that DEC does not have to pay (because Riverside does not want the subsidy) neither that DEC should pay (in the end no one would refuse money, in particular no commercial company). The fact that Riverside does not want to have the subsidy is probably meant as a concession to DEC. This shows that multi-issues cannot be solved one-by-one (piece meal). For that reason the argument support tool may seem to be less useful, because support is provided for just one issue at the time. However, the explanation or support can take the interrelation between issues into account. For
Online negotiation and mediation: Is there room for arguement support tools? Some re... Page 9 of 10 instance, Riverside could support their position that they do not want to receive subsidy amongst others with the statement that this is only so if the installed scrubber will be Rotoblue. In that way the support shows what interdependencies exist between the issues. This type of support may be useful for solving the dispute, but the party providing this type of support has to be careful not to give away information that weakens their position in the negotiation. A last point I want to make is that belief revision is important in negotiation. Due to bargaining, or information provided by the other party, one may start to see things in a different way. This means that it must be possible to withdraw a previous position. The argument support tool must allow this. Withdrawal is an often used move in argument support models. However, in the current version of the argument support tool this is not included yet, so on this point the tool must be adapted. 6. Concluding observations In this paper the possible use of argument support tools in online mediation and in particular online negotiation has been explored. The (preliminary) results of this paper are the following. First, an argument support tool may seem basically adversarial and as such not suited for cooperative negotiation or mediation. This is not true, though. It all depends on the way the argument support tool is used by the parties. To put it extremely, even the opening claims of traditional civil pleading might be used to enter into negotiation in the co-operative ADR sense. The main difference should be that not the plaintiff primarily claims and the defendant denies, but that both parties try to understand the claims of their opponents rather than fighting them. So a procedure does not have to become adversarial due to the tool used, the starting point of the parties is all that counts. To put it the other way around, any existing negotiation tool could be used by parties that fight each other like they do in litigation. In the same way any seemingly adversarial tool can be used in a co-operative way. Whereas litigation is basically adversarial, in negotiation and mediation the motives of the parties make a procedure adversarial or not. Moreover, if parties wanted to play the game in an adversarial way, they probably would not have started an ADR procedure in the first place. Second, the existing argument support tool described in (Lodder & Huygen 2001) should be adapted in several ways in order to be suited for online negotiation. There may be other necessary changes, but in this paper the following has been suggested: allowing withdrawal of statements; providing support must be optional not obligatory. The use of the tool as such must also be optional. Third, an argument support tool is better not used on its own. For instance, numerical issues cannot be represented in a natural way, bargaining ranges cannot be indicated, etc. Therefore the tool preferably is used in combination with existing negotiation and mediation software. The present paper is just a first step towards a thorough analysis of how AI & Law argumentation models, viz. argument support tools, could be applied in online ADR. There is still a lot of work left to do. References Bellucci, E. & J. Zeleznikow (1999), AI techniques for modelling legal negotiation, Proceedings of the Seventh International conference on Artificial Intelligence & Law (ICAIL 99), New York: ACM, p. 108-116. Bench-Capon, T.J.M., T. Geldard & Paul H. Leng (2000), A method for the computational modelling of dialectical argument with dialogue games, AI & Law Vol. 8, Number 2/3, p. 233-254. Bordone (1998), Electronic online dispute resolution: a systems approach potential, problems, and a proposal, Harvard Negotiation Law Review, 175. (see
Online negotiation and mediation: Is there room for arguement support tools? Some... Page 10 of 10 http://eon.law.harvard.edu/h2o/property/jurisdiction/bordoneedit.html). Gordon, T.F. (1995). The Pleadings Game - An artificial Intelligence Model of Procedural Justice. Kluwer, Dordrecht. Hage, J. (2000), Dialectical models in artificial intelligence and law, AI & Law Vol. 8, Number 2/3, p. 137-172. Katsh, E. & J. Rifkin (2001), Katsh & Rifkin, Online Dispute Resolution: Conflict Resolution in Cyberspace, San Fransisco: Jossey-Bass 2001. Leenes, R. (1999), Burden of Proof in Legal Dialogue Systems, in: Proceedings of the Bileta '99 conference: CyberLaw 99, York, 1999 (http://www.bileta.ac.uk/99papers/leenes.html). Lodder, A.R. (1998), On structure and naturalness in dialogical models of argumentation, in J.C. Hage, T.J.M. Bench-Capon, A.W. Koers, C.N.J. de Vey Mestdagh, C.A.F.M. Grütters, Legal Knowledge-based Systems. JURIX: The Eleventh Conference, GNI, Nijmegen, pp. 45-58. Lodder, A.R. (1999), DiaLaw - on legal justification and dialogical models of argumentation, Dordrecht: Kluwer Academic Publishers (Volume 42 of the Law and Philosophy Library), paperback edition 2001. Lodder, A.R. & P.E.M. Huygen (2001), eadr: A simple tool to structure the information exchange between parties in Online Alternative Dispute Resolution, in: B. Verheij, A.R. Lodder, R.P. Loui and A. Muntjewerff (eds.), Legal Knowledge and Information Systems JURIX 2001:The Fourteenth Annual International Conference, Amsterdam, Berlin, etc., IOS Press, p. 117-129. Lodder, A.R. & H.W.K. Kaspersen (2002), edirectives: Guide to European Union Law on E- Commerce Commentary on the Directives on Distance Selling, Electronic Signatures, Electronic Commerce, Copyright in the Information Society, and Data Protection, The Hague, London, New York, Kluwer Law International. Thiessen, E.M. & J.P. McMahon, Beyond Win-Win in Cyberspace, 15 Ohio S. J. on Disp. Resol. 643 (2000), (see http://www.jpmcmahon.com/beyond_win.htm). Ury, W. (1991), Getting past No Negotiating with difficult people, Random house business books (tenth impression 1999).